1. Field of the Invention
The present invention relates to a multi-input multi-output (MIMO) system, and more particularly, to a method and apparatus for allocating transmission power in an MIMO system.
2. Related Art
A multi-antenna technology (or a multi-input multi-output (MIMO) system), which is a technology of improving efficiency and performance of a frequency, has been variously applied to mobile telecommunication and wireless local area network (WLAN). A beamforming technology among the multi-antenna technologies is a technology in which a station transmitting a signal forms and transmits a beam so as to improve packet reception performance of a station receiving the signal using transmission channel information.
In a single user (SU) multi-antenna technology, the beamforming technology allows all transmission streams to be received in a single station, and in a multi-user (MU) multi-antenna technology, the beamforming technology allows the transmission streams to be received in several stations.
In a multi-user multi-antenna scheme, generally, since multiple streams are transmitted simultaneously to a multi-user using the same communication resource, sum capacity of transmission rates capable of being reached in a multi-user environment may be obtained. Therefore, the multi-user multi-antenna scheme has been spotlighted.
In the multi-antenna technology, power appropriate for each transmit antenna is allocated to each transmit antenna, thereby making it possible to increase channel capacity, which is referred to as a power allocation method. Particularly, in the single user multi-antenna technology, it has been known that a power allocation method called a water-filling method is used to allocate high power when a channel is in a good state and allocate low power when the channel is in a bad state, thereby making it possible to obtain optimal channel capacity.
However, in the multi-user multi-antenna technology, in the case in which optimal power is allocated to each user in a situation in which crosstalk is present between users, a non-convex optimization problem occurs. Therefore, a method of solving this problem has been demanded.